[Wien] 1000 atom fix

[Oleg Rubel]

Thank you for the suggestion. I grouped atoms with the same symbol, like "Ga2", in groups of MULT=99. "x sgroup" was able to generate a correct structure file with 60 inequivalent atoms instead of 1200.

PS: We have recently bought HPC cluster that suffices the requirements.
-- 

Oleg Rubel, PhD
Scientist
Thunder Bay Regional Research Institute
290 Munro St
Thunder Bay, ON
P7A  7T1, Canada
Phone: +1-807-7663350
E-mail: rubelo at tbh.net
Homepage: http://www.staff.uni-marburg.de/~rubel/


>>> On 5/26/2009 at 1:33 AM, in message <4A1B7F11.3050004 at theochem.tuwien.ac.at>,
Peter Blaha <pblaha at theochem.tuwien.ac.at> wrote:
> At present WIEN2k is restricted to 999 non-equivalent atoms.
> 
> The calculations I mentioned had inversion symmetry, i.e. about
> 700 non-equivalent atoms, but 1400 atoms/cell.
> 
> Maybe you can start your structure by modifying the struct file
> (by hand or some script) and put atoms with the same Z together
> as "equivalent" atoms (eg. using MULT=48).
> Such a struct file can have more atoms and eventually sgroup or nn
> can rewrite a valid struct file with some symmetry (some equivalent 
> atoms).   If there is NO symmetry, then WIEN2k is limited to 999 atoms.
> 
> Alternatively, just changing nn.f so that it can read more than 1000 
> atoms is fairly trivial (modification of FORMATS 1511, 1012 and probably 
> a few more where "I3" is used.)
> 
> PS: I hope you are aware, that such calculations need a parallel 
> computer with 64-100 cores, infiniband and >50GB memory!
> 
> 
> Oleg Rubel schrieb:
>> Dear Wien2k Users and Developers,
>> 
>> I model antiphase boundary structure in III-V semiconductors. The raw 
> structure I prepared has 1200 atoms and is not optimal from the symmetry 
> point of view. I expect "x sgroup" to generate an optimal structure. The 
> generated structure has 200 atoms, which does not surprise me, but when I 
> look at the structure I see that there are missing atoms. (Everything was 
> fine with 768 atom structure, which was brought by sgroup down to 48 
> independent atoms.) It seems that the number of atoms is limited to 999, at 
> least by sgroup.
>> 
>> I remember that at one of Wien2k workshops Peter spoke about >1000 atoms 
> simulation. I wonder if there is a special patch to unable >1000 atoms?
>> 
>> Thank you in advance.
>> 
>> Sincerely,
>> Oleg
>> 
>> --
>> Oleg Rubel, PhD
>> Scientist
>> Thunder Bay Regional Research Institute
>> 290 Munro St
>> Thunder Bay, ON
>> P7A  7T1, Canada
>> Phone: +1-807-7663350
>> Fax: +1-807-3441948
>> E-mail: rubelo at tbh.net 
>> Homepage: http://www.tbrri.com/~orubel/ 
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at 
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at 
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien